ADA Remediation Playbook — Fix Every Defect
A scan says what is wrong. This says what to do.
Most accessibility tools stop at a list of defects. This one produces a work plan — the kind a project manager can assign to a developer and walk away.
Same engine as the ADA Website Risk Scanner: 14 WCAG 2.2 AA check families covering alt text, form labels, contrast, ARIA, keyboard focus, headings, tables, iframes, touch targets, and media. But the output is fundamentally different.
What you get
For each defect family found on your page:
- Before/after code — the exact HTML pattern to copy. Not a vague instruction; the actual markup, with comments explaining why.
- Effort estimate — how long the batch takes, in developer-hours. Capped at realistic volumes: 5,000 unlabelled images estimates at 4 hours, not 25, because a developer writes a script rather than editing each one by hand.
- Required skill — "Developer" vs "Designer + developer" vs "Content + developer." So you know who to assign.
- Verification step — how to confirm the fix actually works, specific to each defect type.
How the ordering works
Batches are ranked by exposure reduced per hour of developer time, not by raw severity.
A missing lang attribute is a HIGH defect. It also takes five minutes to fix. Missing alt text is CRITICAL — and takes thirty minutes for a handful of images. Fixing the lang attribute first reduces measurable risk faster.
This is not a theoretical distinction. Working strictly by severity leaves easy wins on the table and costs more time for the same risk reduction.
Three phases
| Phase | What goes in | Why it is separate |
|---|---|---|
| 1 — Quick wins | Fastest batches, no design decisions | A developer can finish these today |
| 2 — Core remediation | Substantive work needing testing | The main development effort |
| 3 — Design-dependent | Contrast, touch targets | Needs a designer's sign-off before a developer can proceed |
Phase 3 exists because colour and sizing changes need a decision-maker who is usually not the developer. Without separating them, the whole plan stalls waiting for sign-off on a hex code.
The summary dashboard
At the top of the playbook:
- Total defects found
- Work batches — defects grouped by family
- Phases — how many execution stages
- Estimated effort — total developer-hours
- Exposure score — the same ADA risk score the scanner produces
One glance tells a stakeholder how big the job is.
Who this is for
- Development leads who need to scope accessibility work
- Project managers who need to assign and track remediation
- Agencies delivering accessibility fixes to clients
- Solo developers who want the exact code, not a lecture
What this is NOT
Completing this playbook fixes the defects automated scanning can detect — roughly 30–40% of WCAG 2.2 criteria. The rest require human judgement: whether alt text is meaningful, whether focus order is logical, whether the reading experience makes sense.
This is not a compliance certification. It is a concrete, executable plan that addresses the same defects a plaintiff's automated scan would flag.
Built by ZALTEN — diagnostic agents that find every gap before it finds you.
